A dynamic shuffled differential evolution algorithm for data clustering

Xiang, Wei; Zhu, Ning; Ma, Shoufeng; Meng, Xuelei; An, Mei-qing

doi:10.1016/j.neucom.2015.01.058

“…Actual clustering of the dataset, best obtained clustering of the dataset in both two dimension and three dimension and convergence performance of CMDE is depicted in Figures 1-5 for iris, cancer, thyroid, wine and crude oil dataset respectively. The results of DE/rand/1/bin, DE/best/1/bin, DSDE, ABC, ACO, PSO and PSOAG are collected from paper (Xiang et al, 2015). Considering the results in Table 2 it can be seen that CMDE is significantly better than DE/rand/1/bin and DE/best/1/bin in case of all the datasets.…”

Section: Results and Analysismentioning

confidence: 99%

“…In this section, CMDE for clustering is compared with the most recently proposed partitional algorithm, DSDE (Xiang et al, 2015). This algorithm also has been compared with four evolutionary clustering algorithms such as ACO, PSO, ABC and PSOAG to validate the performance of the proposed CMDE.…”

Section: Resultsmentioning

confidence: 99%

“…To find out the limits of individual feature, Xiang et al (2015) considered the range of each feature of the real dataset and is the same for every cluster. This study considers the upper limit and lower limit of each feature of the dataset in each cluster, i.e., the limits of each feature of one cluster is different from other cluster.…”

Section: Population Presentation and Initialisationmentioning

confidence: 99%

“…PSO, ACO and ABC are coming under the same class of nature inspired population-based optimisation technique. Hence, a comparison among these three natures inspired algorithms along with two recently proposed dynamic shuffled differential evolution (DSDE) (Xiang et al, 2015) algorithm and particle swarm optimisation with age-group topology (PSOAG) (Jiang et al, 2013) algorithm has been made in this literature. Along with these two well-known mutation strategies of DE, such as DE/rand/1/bin and DE/best/1/bin, are also considered for performance comparison.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A cross mutation-based differential evolution for data clustering

Nayak

¹

,

Rout

²

,

Jagadev

³

2017

IJDMMM

2

0

View full text Add to dashboard Cite

Abstract:A cross mutation-based differential evolution (CMDE) approach is proposed here to handle the complexity issue in clustering due to the data uncertainty, overlapping and rapid growth in size of data. In this CMDE, a novel mutation strategy and a centroid rearrangement scheme have been proposed for getting a better and consistent result. CMDE provides optimal cluster centres with minimum intra cluster distance and maximum accuracy percentage. A comparative analysis of the proposed approach with another five population based methods, such as dynamic shuffled differential evolution (DSDE), ant colony optimisation (ACO), artificial bee colony (ABC), particle swarm optimisation (PSO) and particle swarm optimisation with age-group topology (PSOAG) is carried out to justify the better clustering performance of the suggested method. These techniques are applied to seven real datasets and the results reveal the efficacy of the proposed algorithm for clustering in various fields.Keywords: differential evolution; data clustering; intra-cluster distance; centroid rearrangement.Reference to this paper should be made as follows: Nayak, S.K., Rout, P.K. and Jagadev, A.K. (2017) 'A cross mutation-based differential evolution for data clustering ', Int. J. Data Mining, Modelling and Management, Vol. 9, No. 1,

show abstract

“…In order to discuss the result, the proposed algorithm has been compared with some of the recently suggested population-based algorithms such as DSDE (Xiang et al, 2015) and SOAG (Jiang et al, 2013). Also, three algorithms such as ABC, ACO and PSO have also taken into consideration for result evaluation purpose.…”

Section: Results and Analysismentioning

confidence: 99%

A cross mutation-based differential evolution for data clustering

Nayak

¹

,

Rout

²

,

Jagadev

³

2017

IJDMMM

0

View full text Add to dashboard Cite

Abstract:A cross mutation-based differential evolution (CMDE) approach is proposed here to handle the complexity issue in clustering due to the data uncertainty, overlapping and rapid growth in size of data. In this CMDE, a novel mutation strategy and a centroid rearrangement scheme have been proposed for getting a better and consistent result. CMDE provides optimal cluster centres with minimum intra cluster distance and maximum accuracy percentage. A comparative analysis of the proposed approach with another five population based methods, such as dynamic shuffled differential evolution (DSDE), ant colony optimisation (ACO), artificial bee colony (ABC), particle swarm optimisation (PSO) and particle swarm optimisation with age-group topology (PSOAG) is carried out to justify the better clustering performance of the suggested method. These techniques are applied to seven real datasets and the results reveal the efficacy of the proposed algorithm for clustering in various fields.Keywords: differential evolution; data clustering; intra-cluster distance; centroid rearrangement.Reference to this paper should be made as follows: Nayak, S.K., Rout, P.K. and Jagadev, A.K. (2017) 'A cross mutation-based differential evolution for data clustering ', Int. J. Data Mining, Modelling and Management, Vol. 9, No. 1,

show abstract

Forecasting hydropower generation byGFDL‐CM3climate model and hybrid hydrological‐Elman neural network model based on Improved Sparrow Search Algorithm (ISSA)

Wang

¹

,

Wu

²

,

Gholinia

³

2021

Concurrency and Computation

View full text Add to dashboard Cite

One of the most sensitive factors affecting hydropower generation is climate change.The objective of this study is to forecast the hydropower generation under the influence of climate change for the next 50 years. For this purpose, the GFDL-CM3 model is used under three scenarios: RCP2.6, RCP4.5, and RCP8.5 to predict precipitation and temperature. Any change in the inlet flow to the turbine will cause changes in the hydropower output. Therefore, the more accurately the flow is estimated, the hydropower generation is forecasted more accurately. In this research, a hydrological-neural network hybrid model for flow prediction is presented in which the Improved Sparrow Search Algorithm (ISSA) has been used. The results forecasting climate parameters showed that the average annual rainfall, runoff, and evaporation have a downward trend compared to the control period. The minimum and maximum annual temperatures have an increasing trend compared to the control period. To evaluate the hydropower generation Aras dam, the HEC-Ressim reservoir model is utilized. By comparing the average annual hydropower with the control period, it is known that the power generation will decrease in future years. The average annual electricity generated by hydropower under the scenario RCP2.6 and RCP4.5 and RCP8.

show abstract

A dynamic shuffled differential evolution algorithm for data clustering

Cited by 53 publications

References 45 publications

A cross mutation-based differential evolution for data clustering

A cross mutation-based differential evolution for data clustering

A cross mutation-based differential evolution for data clustering

Forecasting hydropower generation byGFDL‐CM3climate model and hybrid hydrological‐Elman neural network model based on Improved Sparrow Search Algorithm (ISSA)

Contact Info

Product

Resources

About